Just a random speculation really – do musicians ever really know what it’s like to listen to us? In my case anyway, you hold violin or viola under your chin – the near ends of the resonating strings are rather nearer to my left ear than my ears are to each other, and there are extra vibrations running though my jawbone to my skull and inner ear without passing through the air at all (albeit in my case somewhat damped by the beard!). So what I hear when playing is certainly not what anyone in an audience hears. ‘Raw’ demo recordings tend to be quite unkind to bowed strings, because of webcam mikes optimised to human speech for one thing, so usually before I release anything I do a treble cut and also noise reduction for the obvious reason! I think these make it a better representation of my playing, but the truth is it most obviously simply makes it sound nicer; it would be very hard for me to tell (and perhaps irrelevant to my purposes) if that light manipulation was actually covering up defects in tone and sound. (I was thinking about this listening back to some demos that had been sent to me, I had overdubbed and then comments had been sent back, making me return to a very unpolished product to follow the responses.) Studio recordings have much more accurate recording systems to start with, but also much more sophisticated sound manipulation systems which are normally used with the goal of the best sound possible, not necessarily the sound most like what could have been heard in the studio during the sessions.
In some ways amplified musicians have it easier – you can get a long lead and stand 20 feet away from your amp while playing and that is a pretty good idea of what the audience would hear. In theory of course you can do it during a gig and find out what they actually hear, though you’ll probably be too busy soloing without tripping over the audience to pay much attention … But acoustic instruments, except if the other sounds around are massively loud, are heard directly even when played amplified, so you still wouldn’t get a comparable impression. (It’s quite rare for the amount of noise on stage to be so loud that I really need monitors to hear myself. This is quite useful given how many gigs I play with amplification but bad or no foldback!)
It’s a bit of a tree falling in the forest with no-one to hear question, but it does reveal that there’s an assumption underlying all the self-promotion all musicians have to do, which is that we sound roughly how we think we do – close enough to it to be worth listening to, close enough that people who hear us won’t say ‘never again’. And in the world of endless statistics and Big Data, the era of prizing scientific evidence above all other means of knowing, that is something it seems we all ultimately have to take on trust.